Effects of alfentanil, preceded by lorazepam, on suppression of haemodynamic and somatic responses to noxious stimuli was studied in patients undergoing CABG. Plasma concentration of alfentanil, somatic and haemodynamic responses were measured at loss of consciousness, tracheal intubation, sternotomy and during multiple application of electrocoagulation. Additional alfentanil was administered i.v. to control unwanted responses. Study 1 (six patients): lorazepam 0.08 mg kg-1 by mouth 1-2 h before operation, alfentanil priming infusion (60 micrograms kg-1 min-1 for 10 min) followed by maintenance infusion (4.5 micrograms kg-1 min-1). With mean plasma alfentanil 1178 (SEM 54) ng ml-1, two patients required supplementary alfentanil to suppress somatic motor responses; one patients required nitroglycerin to control an increase in arterial pressure which was unresponsive to additional alfentanil following sternotomy. Study 2 (13 patients): lorazepam 0.04 mg kg-1 by mouth as premedication; one of three maintenance infusion rates of alfentanil: 5.4 (n = 4), 6.6 (n = 5), or 7.8 (n = 4) micrograms kg-1 min-1, each preceded by a proportional priming infusion. With plasma alfentanil 2181 (62) ng ml-1, somatic motor responses requiring additional alfentanil occurred in nine patients; haemodynamic responses in four of seven patients tested could not be controlled by alfentanil. The highest plasma concentration of alfentanil to prevent response to a stimulus other than tracheal intubation was different between the two studies (P less than 0.05). We conclude that alfentanil alone is insufficient to suppress haemodynamic and somatic motor responses to noxious stimulation during CABG and that the role of premedication is significant.